This paper adapts a thesis submitted for the degree of Master of Philosophy at Cambridge in June 2007. I am most grateful to my supervisor Professor James Diggle for his encouragement and help throughout. Thanks are also owed to my thesis examiners, Professor Richard Buxton and Dr Helen Morales, and to Professor Pat Easterling, David Butterfield and the anonymous readers for Ramus, for their useful comments and criticism.

1. The first study to explicitly focus on this approach is that of I.J.F.|de Jong, ‘Three Off-Stage Characters in Euripides’, Mnemosyne 43 (1990), 1-21CrossRefGoogle Scholar, where she shows how off-stage characters may be used to evoke sympathy. See also Mossman, J., ‘Waiting for Neoptolemus: The Unity of Euripides’ Andromache’, G&R 43 (1996), 143-56Google Scholar; Cox, C.A., ‘Absence and Distance in Euripides’ Andromache: A Social Commentary’, Eos 87 (2000), 197-205Google Scholar; Davidson, J., ‘The Absence of Alkmene in the Trakhiniai’, Hermes 129 (2001), 551-53Google Scholar; Hamstead, S.D., Off-Stage Characters in Greek Tragedy (Diss. Leeds 2005Google Scholar).

2. Evidence for an alternative tradition is found in the Cypria (fr. 27 Davies; not in West) where Polyxena is fatally wounded by Odysseus and Diomedes, and then buried by Neoptolemus. The presence of Neoptolemus may point to some association of Achilles with Polyxena, but this cannot be assumed with any certainty.

3. On the ritual implications of the verb used (σφαγιάζoυσιν) see Casabona, J., Recherches sur le vocabulaire de sacrifices en grec, des origines a la fin de Vipoque classique (Aix-en-Provence 1966), 155-96Google Scholar, and Scafolgio, G., L’Astyanax di Accio: saggio sul background mitografico, testo critico e commento deiframmenti (Brussels 2006), 20Google Scholar with n.13.

4. The on-stage appearance of Achilles is evidenced by the explicit testimony of Apollodorus (de natura deorum 20 = FGrHist 244 F 102a.2), who quotes a few lines spoken by the ghost (fr. 523 Radt). Ps.-Longinus states that Simonides also described such an appearance of Achilles, but it does not necessarily follow that either Sophocles or Simonides dealt specifically with a request for Polyxena.

5. See e.g. Mossman, J., Wild Justice: A Study of Euripides’ Hecuba (Oxford 1995), 45Google Scholar; Mitchell-Boyask, R., Hecuba: Introduction, Translation and Commentary (Newburyport 2006), 30Google Scholar; Sommerstein, A.H., Fitzpatrick, D. and Talboy, T. (eds.), Sophocles: Selected Fragmentary Plays (Warminster 2006), 65f.Google Scholar

6. Dares 27, 34,43; Dictys 3.2-3,4.10-11; Hyginus Fab. 110.

7. Sommerstein et al. (n.5 above), 47; see also Förster, R., ‘Zu Achilleus und Polyxena’, Hermes 18 (1883), 475-78Google Scholar, and Robertson, M., ‘Troilos and Polyxene: Notes on a Changing Legend’, in Descouedres, J.-P. (ed.), Eumousia: Ceramic and Iconographic Studies in Honour of Alexander Cambitoglou (Sydney 1990), 63-70.Google Scholar

8. The text used throughout is that of Diggle, J., Euripidis Fabulae: Tomus I (Oxford 1984Google Scholar). Lines 93-95 were first bracketed by Baier, C., Animadversiones in poetas tragicos Graecos (Diss. Bonn 1874Google Scholar): so also Wilamowitz-Moellendorff, U. von, ‘Lesefriichte’, Hermes 44 (1909), 446-51Google Scholar; Biehl, W., ‘Die Interpolationen in Euripides’ Hekabe v. 59-215’, Philologus 101 (1957), 55-69CrossRefGoogle Scholar, and Textkritik und Formanalyse zur euripideischen Hekabe: ein Beitrag zum Verständnis der Komposition (Heidelberg 1997Google Scholar); Bremer, J.M., ‘Euripides’ Hecuba 59-215: A Reconsideration’, Mnemosyne 24 (1971), 232-50CrossRefGoogle Scholar; Mossman (n.5 above). Diggle, D. Kovacs (in both ‘Coniectanea Euripidea’, GRBS 29 [1988], 115-34Google Scholar, and Children of Heracles; Hippolytus; Andromache; Hecuba [Cambridge MA and London 1995]Google Scholar and Collard, C., Hecuba: with Introduction, Translation and Commentary (Warminster 1991Google Scholar) also delete but with reservation. Arguments for retention are found in Jouanna, J., ‘Réalité et théâtralité du räve; le räve dans l’Hécube d’Euripide’, Ktèma 7 (1982), 43-52Google Scholar; Erbse, H., Studien zum Prolog der euripideischen Tragödie (Berlin 1984CrossRefGoogle Scholar); Brillante, C., ‘Sul prologo dell’ Ecuba di Euripide’, RFIC 116 (1988), 429-47Google Scholar; Gregory, J., Hecuba: Introduction, Text, and Commentary (Atlanta 1999Google Scholar); L. Battezzato in his review of Gregory, , CJ 96 (2000), 223-28Google Scholar; Matthiessen, K., ‘Die Tragödien des Euripides’, Zetemata 114 (Munich 2002Google Scholar). while lines 90f. are certainly to be athetised, the arguments against 93-95 are not so conclusive. Here, therefore, I include them in my discussion. Even if they are deleted, an inconsistency remains between the accounts of Polydorus, Odysseus and the Chorus.

9. Kovacs, D., The Heroic Muse: Studies in the Hippolytus and Hecuba of Euripides (Baltimore and London 1987).Google Scholar

10. Gregory (n.8 above), xxiv-xxix.

11. Michelakis, P., Achilles in Greek Tragedy (Cambridge 2002).Google Scholar

12. Little but not nothing: see 2.2 below, where Achilles’ possible erotic feelings for Polyxena are suggested through the actions of his son Neoptolemus.

13. Easterling, P.E., ‘Philoctetes and Modern Criticism’, ICS 3 (1978), 27-39Google Scholar, at 27; reprinted in Segal, E. (ed.), Oxford Readings in Greek Tragedy (Oxford 1983), 217-28.Google Scholar

14. Collard (n.8 above), 33.

15. Mitchell-Boyask, R., ‘Sacrifice and Revenge in Euripides’ Hecuba’, Ramus 22 (1993), 116-34.CrossRefGoogle Scholar

16. Kovacs (n.9 above), 145 n.58, and Euripidea Altera (Leiden 1996), 63-65Google Scholar, and Gregory (n.8 above), xxix-xxxi, provide the most thorough discussion; see also Conacher, D.J., ‘Euripides’ Hecuba’, AJP 82 (1961), 1-26Google Scholar, at 2; those accepting their arguments include Schlesier, R., ‘Die Bakchen des Hades: dionysische Aspekte von Euripides’ Hekabe’, Métis 3 (1988), 111-35CrossRefGoogle Scholar; Thalmann, W.G., ‘Euripides and Aeschylus’, CA 12 (1993), 126-59Google Scholar; Scodel, R., ‘: Virgin Sacrifice and Aesthetic Object’, TAPA 126 (1996), 111-28Google Scholar, at 121 n.26.

17. For example, Mitchell-Boyask (n.5 above,), 89, calls Hecuba ‘the single most disturbing of the surviving Greek tragic dramas’.

18. The fragments of Polyxena are unhelpful in determining how (if at all) Sophocles portrayed Neoptolemus. Arctinus’ Iliupersis is the earliest text which we know to have featured Polyxena’s sacrifice, but Proclus does not name her killers.

19. LIMC VII.2.347; for discussion of this vase, see Fontinoy, C., ‘Le sacrifice nuptual de Polyxène’, AC 19 (1950), 383-96Google Scholar; Robertson (n.7 above), 64; Mossman (n.5 above), 257.

20. First observed by Tarkow, T.A., ‘Tragedy and Transformation: Parent and Child in Euripides’ Hecuba’, Maia 36 (1984), 123-36Google Scholar, at 125f. (and not Segal, C., ‘Violence and the Other: Greek, Female and Barbarian in Euripides’ Hecuba’, TAPA 120 [1990], 109-31Google Scholar, at 127 n.53, as suggested by Mitchell-Boyask [n.15 above], 122 n.22) but seldom noted since; see only Segal, C., ‘Violence and Dramatic Structure in Euripides’ Hecuba’, in Redmond, J. (ed.), Violence in Drama (Cambridge 1991), 35-46Google Scholar, at 39; Mitchell-Boyask (n.15 above); Thalmann (n.16 above), 142 n.38; Phillippo, S., ‘Family Ties: Significant Patronymics in Euripides’ Andromache’, CQ 45 (1995), 355-71CrossRefGoogle Scholar, at 358. It is true that and are preferable metrically to NεoπτóλεμOς, (with or without synizesis of the first two syllables; see Phillippo, 357f.), but this is the only Euripidean tragedy to refer to Neoptolemus solely as . Forms of NεoπτóλεμOς, though rare, are found in extant tragedy: see Soph. Phil. 4 and 241 (both coupled with ) and Eur. Andr. 14, Tr. 1126, Or. 1655.

21. It appears, however, that in the Ilias parua Priam was killed at the doors of his palace (Pausanias 10.27.2 = llias parua fr. 25 West).

22. Noted by Luschnig, C.A.E., ‘Euripides’ Hecabe: The Time is Out of Joint’, CJ 71.(1976), 227-34Google Scholar, at 227: ‘the violence of Achilles is passed onto his son.’ Tarkow (n.20 above), 126, discussing this idea, posits that ‘a prominent issue of the play, indeed, concerns the existence of inter-generational constancy and its determination of individual characterization and motivation’. Neoptolemus is the prime example: he displays the same traits as, and is motivated solely by the desires of, his father Achilles.

23. This is especially the case given his total absence from the debate over whether or not to sacrifice Polyxena; see 2.3 below. Tarkow (n.20 above), 125f., expresses puzzlement over the choice of Neoptolemus: ‘Despite the insistence of the Greek army that Polyxena be slain, it is not a leading member of the army, or a priest, or a neutral party who accomplishes the deed, but (for reasons never clearly set forth) Neoptolemus.’ He misses the point: being is all the qualification Neoptolemus needs to claim the role.

24. For discussion of (a modern term) in the Athenian marriage ceremony, see Redfield, J., ‘Notes on the Greek Wedding’, Arethusa 15 (1982), 181-201Google Scholar, at 192; Jenkins, I., “‘Is there Life after Marriage?” A Study of the Abduction-Motif in Vase Paintings of the Athenian Wedding Ceremony’, BICS 30 (1983), 137-47Google Scholar, at 139-42; Oakley, J.H. and R.H., Sinos, The Wedding in Ancient Athens (Madison 1993), 32Google Scholar and 137 n.71; Rehm, R., Marriage to Death: The Conflation of Wedding and Funeral Rituals in Greek Tragedy (Princeton 1994), 14-17Google Scholar and 35-40. Several of these critics discuss a black-figure hydria of c.500 BCE (LIMC VII.2.347) which depicts Neoptolemus leading Polyxena towards Achilles’ tomb. Bundrick, S.D., Music and Image in Classical Athens (Cambridge 2005), 184Google Scholar, notes that in iconography the groom sometimes clasps his wife’s hand rather than her wrist, as we find here. Neoptolemus’ act is also recognised as an instance of by Collard (n.8 above), 158, and Gregory (n.8 above), 109.

25. Argued by e.g. Loraux, N., Tragic Ways of Killing a Woman, tr. Forster, A. (Cambridge and London 1987), 37-48Google Scholar; Rabinowitz, N.S., Anxiety Veiled: Euripides and the Traffic in Women (Ithaca and London 1993), 33-36Google Scholar and 54-62; Garrison, E.P., Groaning Tears: Ethical and Dramatic Aspects of Suicide in Greek Tragedy (Leiden and New York1995), 159-61Google Scholar; Hoffman, G., ‘Macarie, Polixène et Iphigénie: les vierges héroïques dans le théâtre d’ Euripide’, in Cavalier, O. (ed.), Silence et fureur: la femme et le mariage en Grèce (Avignon 1996), 249-70Google Scholar, at 257-62.

26. Mitchell-Boyask (n.15 above), 122: ‘Since Neoptolemus has no identity apart from his father, remaining nameless throughout&he is merely a cipher of his father, and it seems as if Achilles himself performs the sacrifice.’

27. For the significance of gold in Hecuba, see Segal, C., ‘Golden Armor and Servile Robes: Heroism and Metamorphosis in Hecuba of Euripides’, AJP 111 (1990), 304-17Google Scholar, at 306, and Ferla, K., Von Homers Achill zur Hekabe des Euripides (Munich 1996), 263Google Scholar with n.115.

28. Gregory (n.8 above), 110.

29. Excluding its specifically medical sense in the Hippocratic corpus, this is the only non-literal usage of νήνεμOς in extant literature up to the time of Euripides.

30. Kovacs’ translation of Hec. 533 (n.8 above, 1995).

31. This does not require one to accept that Achilles stopped the ships by causing the winds to fail; even if the winds are blowing, the ships themselves are in harbour and thus νήνεμoι.

32. See de Jong, I.J.F., Narrative in Drama: The Art of the Euripidean Messenger-Speech (Leiden 1991), 88f.Google Scholar, and Mossman (n.5 above), 142-63.

33. We might contrast Neoptolemus’ absence here with the narrative given by Pausanias (10.25.9 = llias parua fr.18 West) who states that, according to Lesches, Astyanax was thrown from the city walls ‘not by the public decision of the Greeks, but instead Neoptolemus on his own account wished to become his murderer’ (). This explicit opposition of public decree against private decision suggests that Lesches highlighted the lack of public debate over Astyanax’s fate. We also know that it was Neoptolemus himself who threw Astyanax from the walls to his death (Ilias parua frr.29-30 West). Neoptolemus’ absence from the debate in Hecuba is thus all the more striking: once more, a decision must be made over the life of a young Trojan royal, but whereas the epic Neoptolemus was so keen to further his interests that he by-passed the route of public discussion and carried out the murder himself, the tragic Neoptolemus does not even feature in the debate.

34. For discussion of Odysseus’ manipulative rhetoric and the consequent devaluation of Polyxena’s death, see Synodinou, K., ‘Manipulation of Patriotic Conventions by Odysseus in the Hecuba’, Métis 9-10 (1994), 189-96CrossRefGoogle Scholar, and Donzelli, G. Basta, ‘Odisseo nell’ Ecuba di Euripide’, Lexis 19 (2001), 184-97.Google Scholar

35. It has been suggested that Cassandra’s prophetic powers are hinted at when, at Horn. Il. 24.697-706, she alone sees the arrival of Hector’s body and announces its return to the Trojans. It is more likely, however, that the tradition of her powers drew on this scene.

36 For full discussion of Cassandra, see Mason, P.G., ‘Kassandra’, JHS 79 (1959), 80-93CrossRefGoogle Scholar, and Neblung, D., Die Gestalt der Kassandra in der antiken Literatur (Leipzig 1997).CrossRefGoogle Scholar

37. The fullest discussion is that of Thalmann (n.16 above); see also Meridor, R., ‘Eur. Hec. 1035-38’, AJP 96 (1975), 5f.Google Scholar

38. It seems certain that fr.526 Radt comes from such a prophecy. Hickman, R.M., Ghostly Etiquette on the Classical Stage (Cedar Rapids 1938), 49Google Scholar, argues that Cassandra must have delivered it; Sommerstein (n.5 above), 56f., suspects either Cassandra or Polyxena. As such prophecies tend to occur at the very end of tragedies (compare Soph. Phil. 1409-44), I think it unlikely to have been Polyxena, as we would expect her to be dead by this point.

39. Segal, C., ‘The Problem of the Gods in Euripides’ Hecuba’, MD 22 (1989), 9-21Google Scholar, one of the very few critics to comment on Cassandra’s absence, relates these instances to the play’s ‘Dionysiac’ elements (18); see also Schlesier (n.16 above) and Zeitlin, F., ‘Euripides’ Hekabe and the Somatics of Dionysiac Drama’, Ramus 20 (1991), 53-94CrossRefGoogle Scholar, who both, I think, place too much weight on this very slight motif. Indeed Marshall, C.W., A Commentary on Euripides’ Hecuba 658-1295, with an Introduction to the Play as a Whole (Diss. Edinburgh 1992), 145Google Scholar, notes that these terms, when used here of Cassandra, do not appear to have Dionysiac undertones. I am grateful to Dr Marshall for permission to refer to his unpublished thesis.

40. Synodinou, K., ‘Agamemnon in the Hecuba of Euripides: A Case of Submissiveness to, and Misjudgement of, Women’, in (Ioannina 1994), 99-105Google Scholar, at 100.

41. See also King, K.C., ‘The Politics of Imitation: Euripides’ Hekabe and the Homeric Achilles’, Arethusa 18 (1995), 47-66Google Scholar, at 54.

42. First noted by King, K.C., Achilles: Paradigms of the War Hero from Homer to the Middle Ages (Berkeley 1987), 88Google Scholar.

43. It is perhaps noteworthy that Polyxena’s penultimate utterance is to state with confidence that Polydorus is alive and will close Hecuba’s eyes at her own death (430). This statement has a strong vatic tone, and there is a recognisable tradition in Greek literature of those who are about to die suddenly acquiring the ability to predict the future (see Most, G.W., ‘A Cock for Asclepius’, CQ 43 [1993], 96–111Google Scholar, at 108f.). Polyxena’s incorrect statement therefore sounds as though it should be true. Polyxena thus becomes a false and inadequate Cassandra: instead of presenting accurate prophecies which are disbelieved, she creates a untrue prophecy which appears to convince Hecuba. Polyxena’s attempt to bring ‘Cassandra’ on-stage by appropriating her sister’s vatic voice is thus unsuccessful, and we are reminded once more of the inadequacy of purely human information.

44. It is not the case that Agamemnon here retreats, but rather tries to without success: Mercier, C.E., ‘Hekabe’s Extended Supplication (Hec. 752–888)’, TAPA 123 (1993), 149–60Google Scholar, shows that Hecuba continues to hold Agamemnon throughout this exceptionally long supplication, thus intensifying the impression of her desperation.

45. I exclude lines 831f., which are deleted by Matthiae, A.H., Euripidis Tragoediae et Fragmenta (Leipzig 1813Google Scholar); Prinz, R., Euripidis Tragoediae: Hecuba (Leipzig 1883Google Scholar); Murray, G., Euripidis Fabulae: Vol.1 (Oxford 1902Google Scholar); Diggle (n.8 above); Collard (n.8 above); Kovacs (n.8 above, 1995).

46. Grube, G.M.A., The Drama of Euripides (London 1941), 224Google Scholar, concludes that ‘abject supplication can go no further’; Conacher (n.16 above), 22f., finds in Hecuba’s argument an appeal to the ‘bought favours and calculated returns of whores and pimps’; for Buxton, R.G.A., Persuasion in Greek Tragedy (Cambridge 1982), 179Google Scholar, it is ‘demeaning’; Keyser, P.T., ‘Agonizing Hekabe’, ColbyQ 33 (1997), 128–61Google Scholar, writes that ‘nothing can soften the shock or palliate the objection’ (152).

47. Mossman (n.5 above), 126–28; Gregory (n.8 above), 143; Lloyd, M., The Agon in Euripides (Oxford 1992), 96Google Scholar; Kastely, J.L., ‘Violence and Rhetoric in Euripides’ Hecuba’, PMLA 108 (1993), 1036–49CrossRefGoogle Scholar, at 1041–42.

48. Rodríguez, E., ‘Los lechos de Casandra en Troyanas y Hécuba de Euripides y en Alejandro de Licofrón’, EClàs 45 (2003), 25–46Google Scholar, in tracing the motif of Cassandra’s bed in Euripides’ Hecuba and Troades and Lycophron’s Alexandra, notes that Hecuba’s vocabulary here echoes (37), but does not discuss what this reveals about Hecuba’s methods of persuasion.

49. See Riedweg, C., ‘Die Tragödiendichter also Rhetor? Redestrategien in Euripides’ Hekabe und ihr Verhältnis zur zeitgenössischen Rhetoriktheorie’, RhM 143 (2000), 1–32Google Scholar.

50. For discussion of its use in tragedy, see Barrett, W.S., Euripides: Hippolytos (Oxford 1964), 348Google Scholar (on Eur. Hipp. 986f.).

51. Gellie, G., ‘Hecuba and Tragedy’, Antichthon 14 (1980), 30–44CrossRefGoogle Scholar, also sees Hecuba’s argument as a continuation of the Chorus’ narrative at 122–24, but misses the fact that it is a calculated reaction on Hecuba’s part.

52. Scodel, R., ‘The Captive’s Dilemma: Sexual Acquiescence in Euripides’ Hecuba and Troades’, HSCP 98 (1998), 137–54Google Scholar, at 137f.

53. Segal (n.39 above), 18, finds Polymestor a ‘parodistic version of the deus ex machina’ whose prophecies are ‘given not to implement justice, but to compensate for his helplessness and encompass his own revenge against those who have punished his crime’.

54. Meridor, R., ‘Hecuba’s Revenge: Some Observations on Euripides’ Hecuba’, AJP 99 (1978), 28–35Google Scholar, at 33.

55. See also Gellie (n.51 above), 40: ‘Polymestor uses his prophecies as weapons to wound Hecuba and Agamemnon’, and Segal (n.20 above, 1990), 128: ‘He keeps probing for the places he can hurt Hecuba until he finally hits on&the detail that will give her the greatest possible pain.’

56. Cataudella, Q., ‘L’ Ecuba di Euripide’, Dioniso 1 (1939), 118–34Google Scholar, at 127, sees Cassandra as dying ‘per insidia e vendetta di Clitennestra (come i figli di Polimestore per la vendetta di Ecuba)’; Collard, C., ‘Euripides, Hecuba 1056–1106: Monody of the Blinded Polymestor’, in López Férez, J.A. (ed.), Estudios Actuates sobre Textos Griegos (Madrid 1991), 161–73Google Scholar, notes at 169 the symmetry of Polymestor and Hecuba each killing the other’s children.

57. Collard (n.56 above), 169. See also Dunn, F.M., Tragedy’s End: Closure and Innovation in Euripidean Drama (Oxford 1996Google Scholar), who repeatedly terms Polymestor’s prophecy a ‘demonic epiphany’: he sees Polymestor as a figure of authority ‘who is somehow more than human’ (40).

58. See Hogan, J.C., ‘Thucydides 3.52-68 and Euripides’ Hecuba’, Phoenix 26 (1972), 241–57CrossRefGoogle Scholar, at 250, and Segal, C., ‘Tragic Beginnings: Narration, Voice and Authority in the Prologues of Greek Drama’, YCS 29 (1992), 85–112Google Scholar, the latter stating that ‘this prologue [of Polydorus] is indeed supernatural, but its authority is that of human kinship bonds rather than of divine foreknowledge’ (91).

59. Zeitlin (n.39 above), 75.

60. This story also featured in the Mas parua (Ilias parua arg. 4 West).

61. Orban, M., ‘Hecube, Drama Humain’, LEC 38 (1970), 316–30Google Scholar, at 317 n.3.

62. That this is manipulation on the part of Hecuba is particularly apparent given her genealogy in this play, where Euripides makes her the daughter of the Thracian king Cisseus (Hec. 3), as opposed to her Homeric father Dymas; see Gregory, J., ‘Genealogy and Intertextuality in Hecuba’, MP 116 (1995), 389–97Google Scholar, and Schubert, P., ‘L’ Hécube d’ Euripide et la définition de l’étranger’, QUCC 64 (2000), 87–100Google Scholar. Hecuba, despite having Thracian blood, thus presents herself as more civilised than the Thracians in order to win favours from the Greeks.

63. Kovacs (n.8 above, 1988 and 1995), and Euripidea Altera (Leiden 1996Google ScholarPubMed). He is followed by Gregory (n.8 above); for defence of these lines, see Battezzato (n.8 above), 225f., and Mastronarde, D., in his review of Gregory, AJP 123 (2002), 129–32Google Scholar, at 131.

64. West, M.L., ‘Tragica IV, BICS 27 (1980), 9–22Google Scholar.

65. The excision of 441–43 was first proposed by Hartung, followed by, inter alios, Diggle (n.8 above) and Collard (n.8 above); for the most recent arguments for its deletion, along with other instances of interpolated curses in tragedy, see Finglass, P., ‘The Interpolated Curse’, Hermes 134 (2006), 257–68Google Scholar.

66. Thalmann (n.16 above), 132, makes this connection also, but bases his observation upon the description of Helen as Λάκαινα at 441, a line which, as stated, I believe to be interpolated.

67. ή Λάκαινα is used also of Hermione (Eur. Andr. 29 and 486) and Clytaemestra (Eur. IT 806), but as these are both close relatives of Helen, the term serves to remind us of their associations with her.

68. See Dué, C.The Captive Woman’s Lament in Greek Tragedy (Austin 2006), 134fGoogle Scholar.

69. Collinge, N.E., ‘Euripides Hecuba 925–26’, CP 49 (1954), 35f.Google Scholar, at 36; see also Collard, C., ‘The Stasimon Euripides, Hecuba 905–52’, SEJG 31 (1989), 85–97Google Scholar, at 92 (reprinted in Tragedy, Euripides and Euripideans [Exeter 2007], 141–55CrossRefGoogle Scholar).

70. Michelini, A., Euripides and the Tragic Tradition (Madison 1987), 332Google Scholar, sees this simile as evoking the ‘associations of Spartan girls—such as Helene—with beauty and loose sexuality’; I argue that it evokes Helen herself. Collard (n.69 above), 92, sees that this recalls the Λάκαινα of 651, but does not mention Helen; Zeitlin (n.39 above), 75, comments that the Chorus move from this simile to ‘that other Dorian girl—Helen’ at 943–51 but does not see that this is not so much a shift as a continuation of thought; finally Mossman (n.5 above), 90, argues that it ‘recalls in a curiously off-key manner the sympathy for the mourning Greek women in the previous ode’.

71. Marshall, C.W., ‘The Costume of Hecuba’s Attendants’, A Class 44 (2001), 127–36Google Scholar. Quotation at 129.

72. Battezzato, L., ‘Dorian Dress in Greek Tragedy’, ICS 24–25 (1999), 343–62Google Scholar, at 361.

73. We might even suggest—as do several critics—that Polyxena turns into a ‘Helen’ figure: the gesture of baring her breasts at Hec. 558–61 recalls the identical act which Helen was said to have performed in order to stop Menelaus from killing her during the sack (see e.g. Eur. Andr. 629–31).